CN115334167B - An Angle Threshold Adaptive Trajectory Compression Method - Google Patents

Abstract

The invention discloses an angle threshold adaptive trajectory compression method, which relates to the technical field of data compression. Obtain vector map data to extract the original road data and perform grid division to obtain target road data containing multiple grid areas; perform network topology analysis on the target road data, count the number of road intersections contained in the grid area and assign angle thresholds; Calculate the deflection angle value of each track point in the original driving track, store the track points that meet the preset conditions in the cache queue, and obtain the cached driving track; calculate the deflection cumulative value of each track point in the cached driving track, and sample according to the deflection cumulative value and storage to obtain the compressed target driving trajectory of the target vehicle. The above method can effectively compress the trajectory data, and at the same time effectively retain the trajectory data of the vehicle turning, and perform two data compressions according to the deflection angle value and the deflection cumulative value to reduce the amount of trajectory data.

Description

An angle threshold adaptive trajectory compression method

Technical Field

The invention relates to the technical field of data compression, and in particular to an angle threshold adaptive trajectory compression method.

Background Art

Trajectory data includes the location information, time information, speed information, etc. generated by one or more moving objects during their movement in a certain space-time environment. This information is sorted in time series to form the movement trajectory of the moving object. For example, the positioning system used in smart phones can collect and sample the location information of the mobile phone through GPS signals or base station signals, and the vehicle can continuously collect trajectory points during the vehicle's driving process through the on-board positioning system. These collected trajectory data can be sorted by time to draw the movement trajectory data of the moving object over a period of time, which contains rich semantic information.

With the development of 5G and the Internet of Things, GPS equipped with storage hardware has been widely used in various vehicles, thus generating a massive amount of trajectory data. These applications facilitate the tracking, prediction and analysis of vehicle locations. At the same time, as the importance of trajectory data increases, the frequency of tracking points collected by positioning devices is also increasing. They collect location information at a rate of seconds, which leads to problems such as large amounts of trajectory data, high transmission and storage costs, and data redundancy.

Summary of the invention

The purpose of the present invention is to solve the above-mentioned problems in the background technology and to propose a trajectory compression method with adaptive angle threshold.

The purpose of the present invention can be achieved through the following technical solutions:

An embodiment of the present invention provides a trajectory compression method with an angle threshold self-adaptation, the method comprising:

Obtaining vector map data of the current driving area of the target vehicle in a preset map library, extracting original road data from the vector map data and performing grid division to obtain target road data including a plurality of grid areas;

Performing a network topology analysis on the target road data, counting the number of road intersections contained in each grid area, and assigning an angle threshold to the grid area;

During the acquisition of the original driving trajectory of the target vehicle, the deflection angle value of each trajectory point in the original driving trajectory is calculated, and according to the position and deflection angle value of each trajectory point, the trajectory points that meet the preset conditions are stored in a cache queue to obtain a cached driving trajectory;

The deflection accumulation value of each track point in the cached driving track is calculated, and sampling and storage are performed according to the deflection accumulation value to obtain the compressed target driving track of the target vehicle.

Optionally, vector map data of the current driving area of the target vehicle is obtained in a preset map library, original road data is extracted from the vector map data and gridded to obtain target road data containing multiple grid areas, including:

Using the open source map OSM map database, obtaining vector map data of the city or area where the target vehicle is currently traveling, modifying the encoding format of the vector map data to the UTF-8 encoding format, and obtaining vector road data;

Select vector line data from vector road data, select vector line data according to preset field attributes, and export them to original road data in OSM format;

The original road data is rasterized, a new grid is created according to the layer size of the original road data, and the vector map is gridded according to a preset size at the pixel height and width to obtain target road data containing multiple grid areas.

Optionally, the original road data includes at least one of vector line data with field attributes of primary, secondary, tertiary, motoway, and service.

Optionally, a network topology analysis is performed on the target road data, and for each grid area, the number of road intersections contained in the grid area is counted, and an angle threshold is allocated to the grid area, including:

Create a personal geographic database for the target vehicle, create a feature dataset in the personal geographic database and import the target road data, create a topology rule using the feature dataset, remove hanging points and overlapping data in the target road data according to the topology rule, and obtain intermediate target road data;

Using the intermediate target road data to establish a new road data set to obtain road intersection data, for each grid area, according to the different grid fields where the road intersection data is located, counting the number of road intersections contained in the grid area;

For each grid area, calculate the angle threshold of the grid area:

为基于固定角度压缩时角度阈值的取值，l为所述目标道路数据包含的网格区域的数目。Where β _i is the angle threshold of the i-th grid area, c _i is the number of road intersections contained in the i-th grid area,

is the value of the angle threshold when the fixed angle is compressed, and l is the number of grid areas included in the target road data.

Optionally, during the original driving trajectory acquisition process of the target vehicle, the deflection angle value of each trajectory point in the original driving trajectory is calculated, and according to the position and deflection angle value of each trajectory point, the trajectory points that meet the preset conditions are stored in a cache queue to obtain a cache driving trajectory, including:

The driving trajectory data of the target vehicle is collected at a preset frequency to obtain an original driving trajectory;

For each trajectory point of the original driving trajectory, if the trajectory point is a starting trajectory point or an ending trajectory point of the original driving trajectory, the trajectory point is added to a cache queue for caching;

For each trajectory point of the original driving trajectory, if the trajectory point is an intermediate trajectory point of the original driving trajectory, calculating the deflection angle value of the intermediate trajectory point;

For each intermediate track point, the deflection angle value of the intermediate track point is compared with the angle threshold corresponding to the grid area where the intermediate track point is located. If the deflection angle value is greater than the angle threshold, the intermediate track point is selected as a candidate point and added to the cache queue for caching to obtain a cached driving trajectory. Otherwise, the intermediate track point is temporarily retained and discarded after calculating the angle deflection angle value of the next intermediate track point.

For each trajectory point of the original driving trajectory, if the trajectory point is the end trajectory point of the original driving trajectory, the trajectory points stored in the cache queue are obtained and combined to obtain a cache driving trajectory.

Optionally, calculating the deflection angle value of the intermediate trajectory point includes:

Calculate the side length of the triangle formed by the intermediate trajectory point P _cur and its previous trajectory point P _pre and the next trajectory point P _fol as vertices:

和

分别为P_pre和P_cur、P_fol和P_cur以及P_pre和P_fol所构成三角形的边长，Lon_pre和Lat_pre为P_pre的经度和维度；Lon_cur和Lat_cur为P_cur的经度和纬度；Lon_fol和Lat_fol为P_fol的经度和纬度；MLon_i和MLat_i代表经过转换的经纬度数据，i为pre，或者i为cur，或者i为fol，经度为东经时取MLon_i＝Lon_i，经度范围为西经时取MLat_i＝-Lon_i，纬度范围为北纬MLon_i＝90-Lon_i，纬度范围为南纬时取MLat_i＝90+Lat_i；in,

and

are respectively the side lengths of the triangle formed by P _pre and P _cur , P _fol and P _cur, and P _pre and P _fol , Lon _pre and Lat _pre are the longitude and latitude of P _pre ; Lon _cur and Lat _cur are the longitude and latitude of P _cur ; Lon _fol and Lat _fol are the longitude and latitude of P _fol ; MLon _i and MLat i represent the converted longitude and latitude data, i is pre, or i is cur, or i is fol, MLon _i =Lon _i when the longitude is east longitude, MLat _i =-Lon _i when the longitude range is west longitude, MLon _i =90- _{Lon i when} the latitude range is north latitude, and MLat _i =90+Lat _i when the latitude range is south latitude;

Calculate the distance between P _cur and P _pre and P _fol :

Among them, l ₁ represents the distance between P _pre and P _cur ; l ₂ represents the distance between P _fol and P _cur ; l ₃ represents the distance between P _pre and P _fol ; R represents the average radius of the earth, R = 6371.004 km;

Calculate the deflection angle θ _k of P _cur :

Optionally, calculating the deflection cumulative value of each track point in the cached driving track, sampling and storing according to the deflection cumulative value, and obtaining the compressed target driving track of the target vehicle, comprises:

For each track point in the cached driving track, the difference between the deflection angle value and the straight angle is calculated:

代表当前点与平角的差值，θ_k为该轨迹点的偏转角度值；in,

represents the difference between the current point and the straight angle, and θ _k is the deflection angle value of the trajectory point;

Calculate and accumulate the angle differences of N consecutive trajectory points:

将

与

进行比较，若满足

则将所述缓存行驶轨迹中的全部轨迹点存储，得到所述目标车辆压缩后的目标行驶轨迹，否则，将所述缓存行驶轨迹中的全部轨迹点进行1/2的等间隔采样，得到所述目标车辆压缩后的目标行驶轨迹。Determine the angle accumulation threshold according to the N value

Will

and

Compare, if satisfied

All the trajectory points in the cached driving trajectory are stored to obtain the compressed target driving trajectory of the target vehicle; otherwise, all the trajectory points in the cached driving trajectory are sampled at equal intervals of 1/2 to obtain the compressed target driving trajectory of the target vehicle.

The embodiment of the present invention provides a trajectory compression method with adaptive angle threshold, which obtains vector map data of the current driving area of a target vehicle in a preset map library, extracts original road data from the vector map data and performs grid division to obtain target road data containing multiple grid areas; performs network topology analysis on the target road data, and for each grid area, counts the number of road intersections contained in the grid area, and allocates an angle threshold to the grid area; in the process of collecting the original driving trajectory of the target vehicle, calculates the deflection angle value of each trajectory point in the original driving trajectory, stores the trajectory points that meet the preset conditions into a cache queue according to the position and deflection angle value of each trajectory point, and obtains a cached driving trajectory; calculates the deflection cumulative value of each trajectory point in the cached driving trajectory, samples and stores according to the deflection cumulative value, and obtains the compressed target driving trajectory of the target vehicle. The angle threshold is dynamically allocated according to the density of road intersections in the target vehicle driving grid area. In areas with a small number of road intersections, the probability of vehicle turning is relatively small. Using a larger angle threshold can effectively compress the trajectory data. At the same time, using a smaller angle threshold in areas with a large number of road intersections can effectively retain the trajectory data of vehicle turns and the characteristic trajectory points of the vehicle during driving. Data compression is performed twice according to the deflection angle value and the deflection cumulative value to reduce the amount of trajectory data and solve the problems of high transmission and storage costs and data redundancy.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will be further described below in conjunction with the accompanying drawings.

FIG1 is a flow chart of a trajectory compression method with adaptive angle threshold provided by an embodiment of the present invention;

FIG2 is a flow chart of another angle threshold adaptive trajectory compression method provided by an embodiment of the present invention;

FIG3 is a simulation diagram of a trajectory compression method provided by an embodiment of the present invention and a trajectory compression method improved based on DP;

FIG. 4 is a simulation diagram of a trajectory compression method provided by an embodiment of the present invention and a trajectory compression method without introducing a cache queue.

DETAILED DESCRIPTION

The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

The embodiment of the present invention provides a trajectory compression method with an angle threshold adaptive. Referring to FIG1 , FIG1 is a flow chart of a trajectory compression method with an angle threshold adaptive provided by an embodiment of the present invention. The method may include the following steps:

S101, obtaining vector map data of a current driving area of a target vehicle in a preset map library, extracting original road data from the vector map data and performing grid division to obtain target road data including a plurality of grid areas.

S102, performing network topology analysis on the target road data, counting the number of road intersections contained in each grid area, and assigning an angle threshold to the grid area.

S103, during the collection of the original driving trajectory of the target vehicle, the deflection angle value of each trajectory point in the original driving trajectory is calculated, and according to the position and deflection angle value of each trajectory point, the trajectory points that meet the preset conditions are stored in a cache queue to obtain a cached driving trajectory.

S104, calculating the deflection cumulative value of each track point in the cached driving track, sampling and storing according to the deflection cumulative value, and obtaining a compressed target driving track of the target vehicle.

An angle threshold adaptive trajectory compression method provided by an embodiment of the present invention dynamically allocates an angle threshold according to the density of road intersections in a target vehicle driving grid area. In an area with a small number of road intersections, the probability of a vehicle turning is relatively small. A larger angle threshold can be used to effectively compress trajectory data. At the same time, a smaller angle threshold is used for an area with a large number of road intersections, which can effectively retain the trajectory data of vehicle turns and the characteristic trajectory points of the vehicle during driving. The data is compressed twice according to the deflection angle value and the deflection cumulative value, thereby reducing the amount of trajectory data and solving the problems of high transmission and storage costs and data redundancy.

In one implementation, the preset map library may use a self-developed map library, or use an existing open source map library.

In one implementation, the density of the road intersections in the current grid area determines the probability that the vehicle may turn in the grid area. In areas with a small number of road intersections, the probability of a vehicle turning is relatively small, and the trajectory formed by its N consecutive sampling points is more likely to be a straight line. A larger angle threshold can be assigned to the grid area, thereby reducing the number of trajectory points in the cached driving trajectory in the grid area, and ultimately reducing the amount of data on the target driving trajectory. At the same time, for grid areas with a large number of road intersections, the probability of a vehicle turning in the grid area is greater. Using a smaller angle threshold can effectively retain the trajectory data of the vehicle turning and retain the characteristic trajectory points of the vehicle during driving.

In one implementation, during the data compression process, only the cached driving trajectories in the cache queue need to be processed at a time, which can meet the needs of real-time collection and compression, realize online compression of the trajectory, and reduce the transmission and storage pressure of the target driving trajectory.

In one embodiment, step S101 includes the following steps:

Step 1: Use the open source map OSM map database to obtain the vector map data of the city or area where the target vehicle is currently traveling, modify the encoding format of the vector map data to the UTF-8 encoding format, and obtain the vector road data.

Step 2: Select the vector line data in the vector road data, and select the vector line data according to the preset field attributes, and export it as the original road data in OSM format.

Step three, rasterize the original road data, create a new raster according to the layer size of the original road data, divide the vector map into grids according to the preset size at the pixel height and width, and obtain the target road data containing multiple grid areas.

In one implementation, the vector map data may be imported into QGIS software, and the encoding format of the vector map data may be modified to UTF-8 encoding format in Properties-Source Data to obtain vector road data.

In one implementation, a data export tool in an ARCGIS program may be used to export vector line data from vector road data.

In one implementation, the extracted original road data can be rasterized using the ARCGIS program, and the fishnet tool in the data management tool can be used to create a new grid according to the layer size of the road data, and the vector map can be gridded according to the preset size at the pixel height and width to obtain the target road data containing multiple grid areas. The preset size can be set by the technician according to the actual situation and is not limited here. For example, the preset size can be 500m×500m, or the preset size can be 100m×100m, etc.

In one embodiment, the original road data includes at least one of vector line data with field attributes of primary, secondary, tertiary, motoway, and service.

In one embodiment, step S102 includes the following steps:

Step 1: Create a personal geographic database for the target vehicle, create a feature dataset in the personal geographic database and import the target road data, use the feature dataset to create a new topology rule, remove the hanging points and overlapping data in the target road data according to the topology rule, and obtain the intermediate target road data.

Step 2: Use the intermediate target road data to establish a new road data set to obtain road intersection data. For each grid area, count the number of road intersections contained in the grid area according to the different grid fields where the road intersection data is located.

Step 3: For each grid area, calculate the angle threshold of the grid area:

为基于固定角度压缩时角度阈值的取值，l为目标道路数据包含的网格区域的数目。Where β _i is the angle threshold of the i-th grid area, c _i is the number of road intersections contained in the i-th grid area,

is the value of the angle threshold based on fixed angle compression, and l is the number of grid areas contained in the target road data.

In one embodiment, referring to FIG. 2 , based on FIG. 1 , step S103 may include the following steps:

S1031, collecting the driving trajectory data of the target vehicle at a preset frequency to obtain an original driving trajectory.

S1032, for each trajectory point of the original driving trajectory, if the trajectory point is the starting trajectory point or the ending trajectory point of the original driving trajectory, add the trajectory point to the cache queue for caching;

S1033: For each trajectory point of the original driving trajectory, if the trajectory point is an intermediate trajectory point of the original driving trajectory, calculate the deflection angle value of the intermediate trajectory point.

S1034, for each intermediate trajectory point, the deflection angle value of the intermediate trajectory point is compared with the angle threshold corresponding to the grid area where the intermediate trajectory point is located. If the deflection angle value is greater than the angle threshold, the intermediate trajectory point is selected as a candidate point and added to the cache queue for caching to obtain a cached driving trajectory. Otherwise, the intermediate trajectory point is temporarily retained and discarded after the angle deflection angle value of the next intermediate trajectory point is calculated.

S1035: For each trajectory point of the original driving trajectory, if the trajectory point is the end trajectory point of the original driving trajectory, the trajectory points stored in the cache queue are obtained and combined to obtain a cached driving trajectory.

In one embodiment, the deflection angle value of the intermediate track point is calculated in step S1033 as follows:

Step 1: Calculate the side length of the triangle formed by the intermediate trajectory point P _cur and its previous trajectory point P _pre and the next trajectory point P _{fol as} vertices:

和

分别为P_pre和P_cur、P_fol和P_cur以及P_pre和P_fol所构成三角形的边长，Lon_pre和Lat_pre为P_pre的经度和维度；Lon_cur和Lat_cur为P_cur的经度和纬度；Lon_fol和Lat_fol为P_fol的经度和纬度；MLon_i和MLat_i代表经过转换的经纬度数据，i为pre，或者i为cur，或者i为fol，经度为东经时取MLon_i＝Lon_i，经度范围为西经时取MLat_i＝-Lon_i，纬度范围为北纬MLon_i＝90-Lon_i，纬度范围为南纬时取MLat_i＝90+Lat_i。in,

and

are respectively the side lengths of the triangle formed by P _pre and P _cur , P _fol and P _cur, and P _pre and P _fol , Lon _pre and Lat _pre are the longitude and latitude of P _pre ; Lon _cur and Lat _cur are the longitude and latitude of P _cur ; Lon _fol and Lat _fol are the longitude and latitude of P _fol ; MLon _i and MLat i represent the converted longitude and longitude data, i is pre, or i is cur, or i is fol, MLon _i =Lon _i when the longitude is east longitude, MLat _i =-Lon _i when the longitude range is west longitude, MLon _i =90- _{Lon i when} the latitude range is north latitude, and MLat _i =90+Lat _i when the latitude range is south latitude.

Step 2: Calculate the distance between P _cur , P _pre and P _fol :

Among them, l ₁ represents the distance between P _pre and P _cur ; l ₂ represents the distance between P _fol and P _cur ; l ₃ represents the distance between P _pre and P _fol ; R represents the average radius of the earth.

Step 3: Calculate the deflection angle θ _k of P _cur :

In one implementation, R=6371.004 km.

In one embodiment, step S104 may include the following steps:

Step 1: For each track point in the cached driving track, calculate the difference between the deflection angle value and the straight angle:

代表当前点与平角的差值，θ_k为该轨迹点的偏转角度值。in,

Represents the difference between the current point and the straight angle, _{and θk} is the deflection angle value of the trajectory point.

Step 2: Calculate and accumulate the angle differences of N consecutive trajectory points:

将

与

进行比较，若满足

则将缓存行驶轨迹中的全部轨迹点存储，得到目标车辆压缩后的目标行驶轨迹，否则，将缓存行驶轨迹中的全部轨迹点进行1/2的等间隔采样，得到目标车辆压缩后的目标行驶轨迹。Step 3: Determine the angle accumulation threshold based on the N value

Will

and

Compare, if satisfied

All the trajectory points in the cached driving trajectory are stored to obtain the compressed target driving trajectory of the target vehicle. Otherwise, all the trajectory points in the cached driving trajectory are sampled at equal intervals of 1/2 to obtain the compressed target driving trajectory of the target vehicle.

例如，当N＝5时，

当N＝10时，

等等。In one implementation, the angle accumulation threshold can be selected according to the N value

For example, when N=5,

When N = 10,

etc.

反映当前轨迹点的偏转程度，

越大表示当前车辆行驶中转弯的幅度也越大，

反映当前连续的N个轨迹点的偏转程度，

越大表示连续的N个轨迹点组成的轨迹的转弯的幅度也越大。当

可以将该连续的N个轨迹点全部存储，有效地保留车辆转弯的轨迹数据，保留车辆行驶中的特征轨迹点，否则，连续的N个轨迹点组成的轨迹的转弯的幅度不大，即可对N个轨迹点进行1/2的等间隔采样，压缩轨迹数据，降低目标行驶轨迹的数据量。In one implementation,

Reflects the deflection degree of the current trajectory point.

The larger the value, the greater the turning range of the vehicle.

Reflects the deflection degree of the current N consecutive trajectory points,

The larger the value, the greater the turning range of the trajectory composed of N consecutive trajectory points.

The continuous N trajectory points can all be stored to effectively retain the trajectory data of the vehicle turning and the characteristic trajectory points of the vehicle during driving. Otherwise, the turning amplitude of the trajectory composed of the continuous N trajectory points is not large, and the N trajectory points can be sampled at 1/2 equal intervals to compress the trajectory data and reduce the data volume of the target driving trajectory.

The effect of the present invention can be further illustrated by the following simulation experiment.

Simulation experiment conditions: The simulation experiment platform of the present invention is: the processor is Intel i7-9750 CPU, the main frequency is 2.6GHz, and the memory is 16GB.

The software platforms for the simulation experiment of the present invention are: Windows 10 system, IntelliJ IDEA 2020.1.1 and ArcMap 10.6.

The comparison method used in the simulation is an improved DP algorithm proposed by Liu J, Li H, and Yang Z in their paper “Adaptive Douglas-Peucker Algorithm With Automatic Thresholding for AIS-Based Vessel Trajectory Compression” (IEEE Access, 2019, 7: 150677-150692.).

缓冲队列长度N＝5。The trajectory data for the simulation experiment comes from the GeoLife project of GeoLife GPS Trajectories, which records the movement information of users during various outdoor activities, using the WGS 84 coordinate system. The map data comes from the Beijing vector map data provided by OSM map. The simulation process uses the trajectory data of user No. 030 in the trajectory data on February 10, 2009. The driving range is divided into 500m×500m grids, and the cumulative angle threshold is set to

The buffer queue length N=5.

Simulation content and result analysis:

Simulation 1: Under the above simulation conditions, the trajectory data of user No. 030 on February 10, 2009 are compressed using the present invention and the improved DP algorithm. The results are shown in FIG3 , where:

Figure 3(a) shows the result of trajectory data compression by the improved DP algorithm;

FIG3( b ) shows the result of trajectory data compression based on the angle threshold adaptive trajectory compression method of the present invention.

As can be seen from FIG3 , the improved DP algorithm performs compression based on all trajectory data, and the obtained data results are evenly distributed. The compression results of the present invention have higher compression efficiency while maintaining uniform compression results. In addition, during the data compression process, the present invention only requires N trajectory point data at a time to run, which can meet the needs of real-time acquisition and compression and realize online compression of the trajectory. However, the improved DP algorithm needs to obtain the complete trajectory at one time. The existing technology cannot meet the needs of online compression and thus will generate more transmission and storage pressure.

Simulation 2: Under the above simulation conditions, the effect of the cache queue proposed by the present invention is verified, and the result is shown in FIG4 , where:

FIG4( a ) shows the compression result of trajectory data by the trajectory compression method with angle threshold adaptiveness without introducing a cache queue;

FIG. 4( b ) shows the compression result of trajectory data by the trajectory compression method with adaptive angle threshold of the buffer queue introduced in the present invention.

As can be seen from Figure 4, the method without introducing the cache queue performs poorly in compressing the vehicle trajectory in the straight-ahead situation, and continuous and dense trajectory segments that are approximately straight lines appear. The present invention, by introducing the cache queue, improves the compression efficiency of the vehicle trajectory in the straight-ahead section while keeping the algorithm complexity unchanged.

The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims (7)

1. A method of track compression with adaptive angle threshold, the method comprising:

acquiring vector map data of a current running area of a target vehicle in a preset map library, extracting original road data from the vector map data, and performing grid division to obtain target road data comprising a plurality of grid areas;

performing network topology analysis on the target road data, counting the number of road intersection points contained in each grid area, and distributing an angle threshold value for the grid area;

calculating deflection angle values of all track points in the original running track during the original running track acquisition process of the target vehicle, and storing the track points meeting preset conditions into a cache queue according to the positions and the deflection angle values of all track points to obtain a cache running track;

and calculating a deflection accumulation value of each track point in the cache running track, and sampling and storing according to the deflection accumulation value to obtain the target running track of the target vehicle after compression.

2. The method for compressing a track with adaptive angle threshold according to claim 1, wherein obtaining vector map data of a current driving area of a target vehicle in a preset map library, extracting original road data from the vector map data and performing meshing to obtain target road data including a plurality of meshing areas, comprises:

using an open source map OSM map database to obtain vector map data of a current running city or running area of a target vehicle, and modifying the coding format of the vector map data into a UTF-8 coding format to obtain vector road data;

vector line data in the vector road data are selected, and the vector line data are selected according to preset field attributes, so that original road data in an OSM format are derived;

and carrying out rasterization processing on the original road data, creating a grid according to the layer size of the original road data, and carrying out grid division on a vector map at the pixel height and width according to preset dimensions to obtain target road data comprising a plurality of grid areas.

3. The method of claim 2, wherein the raw road data comprises at least one of vector line data with a field attribute of primary, secondary, tertiary, motoway, service.

4. The method for track compression with adaptive angle threshold according to claim 1, wherein performing network topology analysis on the target road data, counting the number of road intersections included in each grid area, and assigning an angle threshold to the grid area comprises:

newly establishing a personal geographic database of the target vehicle, establishing an element data set in the personal geographic database, importing the target road data, creating a topology rule by using the element data set, and removing hanging points and overlapping data in the target road data according to the topology rule to obtain intermediate target road data;

establishing a new road data set by using the intermediate target road data to obtain road intersection point data, and counting the number of road intersection points contained in each grid area according to the difference of grid fields in which the road intersection point data are located for each grid area;

for each grid region, calculating an angle threshold for the grid region:

wherein ,β_i An angle threshold, c, for the ith grid region _i For the number of road intersections included in the ith mesh region,

for the value of the angle threshold value based on fixed angle compression, l is the number of grid areas contained in the target road data.

5. The track compression method of claim 1, wherein in the process of acquiring the original running track of the target vehicle, calculating the deflection angle value of each track point in the original running track, and storing the track points meeting the preset condition into a cache queue according to the positions and the deflection angle values of each track point to obtain the cache running track, comprising:

acquiring the driving track data of the target vehicle at a preset frequency to obtain an original driving track;

for each track point of the original running track, if the track point is a starting track point or an ending track point of the original running track, adding the track point into a cache queue for caching;

for each track point of the original running track, if the track point is a middle track point of the original running track, calculating a deflection angle value of the middle track point;

comparing the deflection angle value of each intermediate track point with an angle threshold value corresponding to the grid region where the intermediate track point is located, selecting the intermediate track point as a candidate point and adding the candidate point into a cache queue for caching if the deflection angle value is larger than the angle threshold value, so as to obtain a cache driving track, otherwise, temporarily retaining the intermediate track point, and discarding the intermediate track point after the angle deflection angle value of the next intermediate track point is calculated;

and aiming at each track point of the original running track, if the track point is the ending track point of the original running track, acquiring the track points stored in the cache queue, and combining to obtain the cache running track.

6. The method of claim 5, wherein calculating the deflection angle value of the intermediate track point comprises:

calculate the intermediate track point P _cur With the previous trace point P _pre And the latter locus point P _fol Side length of triangle formed by vertexes:

wherein ,

and

Respectively P _pre and P_cur 、P _fol and P_cur P _pre and P_fol Side length of triangle formed and Lon _pre and Lat_pre Is P _pre Longitude and latitude of (a); lon (Lon) _cur and Lat_cur Is P _cur Longitude and latitude of (a); lon (Lon) _fol and Lat_fol Is P _fol Longitude and latitude of (a); MLon _i and MLat_i Representing the converted longitude and latitude data, i is pre, i is cur, i is fos, and longitude is the east longitude time MLon _i ＝Lon _i Taking MLat with longitude range of Western time _i ＝-Lon _i The latitude range is North latitude MLon _i ＝90-Lon _i Taking MLat when latitude is in south latitude _i ＝90+Lat _i ；

Calculation of P _cur And P _pre and P_fol Distance between:

wherein ,l₁ Represents P _pre and P_cur Is a distance of (2); l (L) ₂ Represents P _fol and P_cur Is a distance of (2); l (L) ₃ Represents P _pre and P_fol Distance of (2)The method comprises the steps of carrying out a first treatment on the surface of the R represents the average radius of the earth, taking r= 6371.004km;

calculation of P _cur Deflection angle value theta of (2) _k ：

7. The track compression method of claim 1, wherein calculating a deflection accumulation value of each track point in the buffered running track, sampling and storing according to the deflection accumulation value, and obtaining the target running track after the target vehicle is compressed, comprises:

calculating a difference value between a deflection angle value and a flat angle for each track point in the buffer running track:

wherein ,

representing the difference between the current point and the flat angle, θ _k A deflection angle value for the track point;

calculating and accumulating angle differences of N continuous track points:

determining an angle accumulation threshold from the N value

Will->

And->

Comparing if it meets->

And storing all track points in the cache running track to obtain a target running track after the target vehicle is compressed, otherwise, sampling all track points in the cache running track at 1/2 equal intervals to obtain the target running track after the target vehicle is compressed. />

Images